Mapping Relationship Discovery of Multidimensional Architectures in Autonomous Transportation System Based on Text-Matching Model

Table 8

The performance of different pretraining models.

No	Text-matching model	Precision (%)	Recall (%)	F1 (%)
1	Siamese-BERT-Bi-LSTM	87.03	87.26	87.15
2	Siamese-ALBERT-Bi-LSTM	87.28	86.18	86.73
3	Siamese-roberta-Bi-LSTM	87.02	86.72	86.87
4	Siamese-BERT-wwm-Bi-LSTM	87.30	88.35	87.82

Bold values refer to the maximum prediction performance in each circumstance.